From: email@example.com (Robert Richter)
Subject: Re: Questions on building a battery charger
Date: 27 Nov 2002 07:56:40 -0800
NNTP-Posting-Date: 27 Nov 2002 15:56:41 GMT
"Dave VanHorn" wrote in message news:...
> . The unit would be modified
> > so the adapter plug went directly to the battery. I also need to know
> > where to buy AA batteries with welded on tabs.
> Your local batteries plus will weld up any pack you like, if you buy the
> cells from them.
> > I am capable of taking 1600mAH and dividing by charge time to get
> > current, but my question was, "what is the efficiency of NiMH when
> > charging?"
> I've not seen a figure on this that I'd call solid. I'd say 120% is about
> right, but it will vary with charge speed, lower at high rates.
> > As far as the Atmel Tiny-26, I am building a one-of-a-kind unit, so
> > assembly language for just one product is not feasible--not to mention
> > that I don't have a programmer or assembler for this chip.
> www.atmel.com Studio 3.5 is free, and is your assembler and simulator.
> FBPRG16 is a freeware programmer, needs six wires off your parallel port.
> > I am not trying to go into mass production, but simply building a one-of-a
> > unit. The application is one where removal of the batteries is not an
> > option.
> I hear ya. :)
> The 712 isn't bad, except for the high trickle current.
> If you can figure a way to cut that down, you're home free.
> It's a bit clunky, but maybe a FET that's turned on when you have input
> power, that drains off most of the trickle charge through a resistor?
> > As far as C/300, some people on the web say C/10 doesn't hurt anything
> > no matter how long it is left on the charger, and others think that
> > any over charge slowly ruins the battery.
> C/10 is fine for NICADs, but it will push NIMH batteries into charge
> C/300 is where you want to be. See McGraw Hill=s handbook of batteries.
> > Is it a problem to use the
> > trickle charge current of the '712 if the unit is used during the day
> > and plugged in each night to be topped off?
> You might get away with it.
If C/10 is bad, is there any reason I can't put in a shunt resistor
across the battery to adjust the trickle current? The circuit already
has a diode going to the battery, so the worst it could do is not
trickle charge, but it won't discharge the battery. The shunt
resistor would go before the diode, straight off the transistor and go
to the sening resistor. Is it a problem if this resistor is sized so
that no trickling occurs? Isn't there sufficient overcharge to
equalize the cells before charge ends? Is oxygen only recombined when
charging goes on, so an abrupt charge termination without trickling
would leave the cell under pressure, or will the oxygen recombine